diff options
| author | Andrew Kelley <andrew@ziglang.org> | 2022-02-06 20:23:40 -0700 |
|---|---|---|
| committer | Andrew Kelley <andrew@ziglang.org> | 2022-02-06 20:23:40 -0700 |
| commit | 65b6faa0485253b284f7a63601dc7d0f5858515a (patch) | |
| tree | 33e527cab9dc08c381046ec6199424fc3a6e24dd /lib/std/debug.zig | |
| parent | 3bcce5f6d1f48e20dd177a7e440ddea1c451e779 (diff) | |
| download | zig-65b6faa0485253b284f7a63601dc7d0f5858515a.tar.gz zig-65b6faa0485253b284f7a63601dc7d0f5858515a.zip | |
Sema: avoid `@intToFloat` for f80 which breaks on non-x86 targets
Currently Zig lowers `@intToFloat` for f80 incorrectly on non-x86
targets:
```
broken LLVM module found:
UIToFP result must be FP or FP vector
%62 = uitofp i64 %61 to i128
SIToFP result must be FP or FP vector
%66 = sitofp i64 %65 to i128
```
This happens because on such targets, we use i128 instead of x86_fp80 in
order to avoid "LLVM ERROR: Cannot select". `@intToFloat` must be
lowered differently to account for this difference as well.
Diffstat (limited to 'lib/std/debug.zig')
0 files changed, 0 insertions, 0 deletions